Managing HP Serviceguard for Linux, Sixth Edition, August 2006

Planning and Documenting an HA Cluster
Package Configuration Planning
Chapter 4 111
and will be able to fail over automatically to another
node. If AUTO_RUN is set to NO, the package will not
start up automatically when the cluster starts running,
and will not be able to fail over automatically to
another node.
Default is YES, which allows a package to start up
normally on an available node. Enter YES or NO in the
package ASCII file.
NOTE Serviceguard disables AUTO_RUN automatically
whenever the command cmhaltpkg is executed. The
user needs to re-enable AUTO_RUN using the cmmodpkg
-e <pkg> command.
NODE_FAIL_FAST_ENABLED
In the event of the failure of the control script itself or
the failure of a subnet, if this parameter is set to YES,
Serviceguard will issue a TOC on the node where the
control script fails.
Enter YES or NO. The default is NO.
If NODE_FAIL_FAST_ENABLED is set to YES, the node
where the package is running will be halted if one of
the following failures occurs:
• A package subnet fails and no backup network is
available
• The halt script does not exist
• Serviceguard is unable to execute the halt script
•The halt script times out
However, if the package halt script fails with “exit 1”,
Serviceguard does not halt the node, but sets
NO_RESTART for the package, which causes package
switching (AUTO_RUN) to be disabled, thereby
preventing the package from starting on any adoptive
node.
RUN_SCRIPT and HALT_SCRIPT